Stellar Occultation

What a stellar occultation actually is

An occultation is the simplest of geometric coincidences: a foreground body in our solar system passes between us and a background star, briefly blocking the star's light. The Moon does this several times per night; planets do it occasionally; asteroids do it thousands of times per year, mostly to faint stars; trans-Neptunian objects do it rarely but with extraordinary scientific yield. The event lasts seconds to minutes, the light-curve is digitised at millisecond cadence, and from those few seconds of data come measurements of object size, shape, atmosphere, rings, and binarity that would otherwise require a spacecraft visit.

The deep magic of the technique is that it converts a hopeless angular-resolution problem into a tractable time-resolution problem. Resolving 1 km of structure on an asteroid 1 AU away requires angular resolution of 7 nanoradians — far below what any ground or space telescope can do. But that 1 km of structure crosses the line of sight in 50 ms at typical sky speeds, and a 1-millisecond GPS-timed video camera resolves it directly. By exploiting the foreground body's known relative motion, occultations achieve effective resolutions a thousand times finer than any imaging system.

The occulting body need not be small. The Moon occults stars at a rate of dozens per night during its passage along the ecliptic; planets occult background stars; asteroid occultations are rarer but predictable; TNO occultations are rare and difficult, requiring dense observer networks across continents. Each class has its own light-curve signature and its own scientific payoff.

The geometry: shadow paths and how to be in one

An occultation creates a shadow on Earth as the foreground body moves. The shadow path is a strip across the planet's surface, typically a few hundred to a few thousand kilometres wide for asteroids, a few thousand for TNOs, and most of a hemisphere for the Moon and planets. To observe the event, an observer must be physically inside the shadow path during the predicted seconds-long passage.

The shadow path's predicted location depends on (a) the precise sky position of the background star and (b) the precise ephemeris of the foreground body. Pre-Gaia, star positions were known to 50–100 mas, asteroid ephemerides to ~30 mas — combined uncertainties produced shadow path errors of hundreds of km, often missing the predicted observer entirely. With Gaia DR3 (1 mas systematic, 0.05 mas typical statistical) and modern asteroid astrometry, prediction accuracy has improved by an order of magnitude. Real-time updates from monitoring observatories shift the predicted path within hours of the event itself.

The shadow speed across the Earth is set by the foreground body's relative motion: roughly 5–50 km/s for asteroids, ~25 km/s for the Moon, ~20 km/s for outer planets and TNOs. At typical asteroid speeds the shadow crosses Earth in 1–10 minutes from end to end, but any single observer in the path sees the star drop only for the seconds during which the asteroid actually covers the line of sight.

Reading the light-curve

The shape of an occultation light-curve is a fingerprint of the foreground body's structure.

Feature	Light-curve signature	Source
Bare solid limb	Sharp drop in ~10 ms (diffraction-broadened)	Asteroid, airless TNO
Atmosphere	Gradual drop over 10–100 s	Refraction in increasingly dense layers
Rings	Sharp deeps before/after main occultation	Uranus, Neptune, Chariklo, Haumea
Satellites	Multiple isolated events on same path	Asteroid binaries, TNO satellites
Limb roughness	Edges differ by ms between observers	Surface topography, irregular shape
Diffraction fringes	Oscillation at ingress/egress	Wave-optical fringes near the limb

The duration of each observer's "drop" gives a chord length across the silhouette of the foreground body. If many observers are inside the shadow path with millisecond-accurate timing, the collected chords reconstruct the silhouette geometrically — typically to better than 1 km accuracy on a 100-km asteroid. Photometric depth gives the relative brightness of star and target. Any deviation from sharp ingress/egress betrays atmosphere or rings.

Lunar occultations

The Moon is the most prolific occulter. As it moves at ~25 km/s across the celestial sphere, it occults hundreds of bright stars per night in a band ±5° around its path. Lunar occultations have been observed since antiquity (the first recorded was Aristarchus of Samos's lunar occultation of Mars, 357 BCE), but the technique became scientifically powerful in the 20th century when high-speed photometry could time the disappearance event.

The lunar limb is sharp on a scale of a few metres. The diffraction pattern of starlight at the limb has fringes spaced by ~13 m (Earth-Moon distance, optical wavelength), corresponding to ~30 ms in time at the lunar limb's apparent motion. High-speed photometry of the diffraction fringe pattern provides an extraordinary tool: angular resolution of milliarcsecond scale, used historically to measure the size of nearby giant stars (Antares, Aldebaran's companion) and to discover unsuspected close binaries — the "lunar occultation method" was the highest angular-resolution technique available before space telescopes and adaptive optics.

Planetary rings and atmospheres

The most celebrated occultation discoveries are the ring systems of Uranus, Neptune, and the Centaur Chariklo. Each was found by accident or near-accident during an observation aimed at something else.

Uranus, 10 March 1977. James Elliot and colleagues flying the Kuiper Airborne Observatory observed the occultation of SAO 158687 by Uranus. Their goal was to characterise Uranus's atmosphere from the gradual ingress profile. The plan worked, but they also recorded five sharp deeps in the light-curve before and after the main event, symmetric about it. Five rings, never before known. The discovery of the Uranus ring system from a single 35-minute observation is a textbook example of the technique's power.

Neptune, 22 July 1984. Two French teams observed an occultation of star 28 Sgr by Neptune. They saw partial deeps before and after the main event — but on only one side of the planet's pass. The asymmetric pattern suggested ring arcs (incomplete rings) rather than full rings. Voyager 2's flyby in 1989 confirmed the arcs as fragmentary structures within otherwise full but very faint rings.

Chariklo, 3 June 2013. Felipe Braga-Ribas and colleagues observed an occultation of star UCAC4 248-108672 by the Centaur 10199 Chariklo. Two sharp drops appeared symmetrically before and after the main event — Chariklo has rings. Chariklo is a 250-km-diameter object orbiting between Saturn and Uranus; the discovery established that small-body ring systems exist outside the giant planets. Two rings (C1R and C2R) with widths of 7 and 3 km were resolved.

Pluto's atmosphere, 9 June 1988. An occultation observed from the Kuiper Airborne Observatory and ground-based stations showed the dropoff in starlight was not sharp — it was gradual, with the characteristic exponential profile of refraction in an atmosphere. The first detection of Pluto's nitrogen atmosphere; subsequent occultations through the 1990s and 2000s tracked the atmosphere doubling in pressure as Pluto approached perihelion (1989).

Worked example: timing an asteroid occultation

Suppose 9th-magnitude TYC 1234-5678-1 will be occulted by 89-Julia (asteroid #89, an S-type with diameter ~145 km) on 2025-08-15 at 03:42:30 UT. Predicted shadow path runs from northern Spain through southern France to northern Italy. The relative sky speed of the asteroid against the star is 14 km/s. Compute the maximum possible occultation duration and ingress/egress diffraction scale.

Asteroid diameter D = 145 km
Relative sky speed v = 14 km/s
Maximum chord = D = 145 km (central crossing)
Maximum duration t_max = D / v = 145 / 14 = 10.4 s

An observer near the centre of the shadow path will see the star vanish for ~10 seconds. Observers near the edge see shorter chords (a star at the very edge of the shadow path may have only ~1 s of disappearance). Multiple observers' timing recovers the asteroid silhouette.

Diffraction limit (Fresnel scale) at asteroid distance 2.6 AU = 3.89 × 10¹¹ m:
F = √(λ · d / 2)
  = √(550 × 10⁻⁹ · 3.89 × 10¹¹ / 2)
  = √(107000) m
  ≈ 327 m

Ingress and egress are sharp on length scales of ~327 m, which at 14 km/s corresponds to:

t_diffraction = 327 / 14000 = 0.023 s = 23 ms

So the photometric drop transitions from full starlight to zero in ~23 ms. A video camera at 60 fps (16.7 ms cadence) can just resolve this; a high-speed CMOS at 1000 fps fully resolves the diffraction edge. The transition's exact shape depends on whether the asteroid limb is sharp (a single Fresnel-pattern fringe) or has structure: surface topography of more than a few hundred metres causes detectable departures from the simple model.

Combined timing budget for shape recovery:
  N observers needed: ~10 across the shadow path
  Each observer's chord uncertainty: ~σ_t · v ≈ 0.005 s · 14 km/s ≈ 70 m
  → Asteroid shape recovered to ~70 m on a 145 km body

This is more than 100× the angular resolution of the most powerful adaptive-optics imager directed at the same object, achieved with consumer-grade equipment plus careful timing. The advantage is enormous and explains why occultations remain a productive technique even in an era of ELTs.

The IOTA network and amateur science

The International Occultation Timing Association (IOTA) coordinates global occultation observations. Founded in 1975, it operates a hybrid amateur-professional network: predictions distributed publicly, observers in dozens of countries, results pooled and analysed centrally. The network observes hundreds of asteroid occultations per year, plus all major TNO and Centaur events. Membership is open and the technical bar is modest — a 10-inch telescope, a video camera with GPS time-stamping, and a clear sky from the predicted path.

The professional payoff of amateur networks is large. Gaia astrometry has made predictions sufficiently accurate that ~50% of attempted asteroid occultations succeed (vs ~20% pre-Gaia). Each successful event generates dozens of timing chords. Two-thirds of the asteroid silhouettes in the JPL "Asteroid Pole and Shape Database" derive at least partly from IOTA-coordinated occultations.

Where stellar occultations show up

• Solar-system body characterisation. Sizes and shapes of more than 1500 asteroids have been measured to better than 5% precision through occultations. Particularly for small asteroids (≲ 50 km), occultations are the only resolved-shape data available without spacecraft. Recent NEA threat assessments rely heavily on occultation-derived diameters.
• Pluto's atmospheric history. Sequential occultations from 1988 to 2025 have tracked Pluto's nitrogen atmospheric pressure: rising from ~3 μbar in 1988 to ~10 μbar at perihelion (1989), staying high during the 2015 New Horizons epoch (the spacecraft's REX experiment confirmed 11.5 μbar), and now slowly declining as Pluto recedes from the Sun. The pressure doubles or halves on multi-year timescales as the surface ices condense or sublime.
• TNO discoveries. The MASCARA survey, OSSOS, and dedicated TNO-occultation campaigns have revealed satellites, ring systems (Haumea, possibly Quaoar), and refined diameters of dozens of trans-Neptunian objects. The Lucky Star programme alone has contributed sizes for ~50 large TNOs.
• Stellar diameters via lunar occultation. Pre-AO and pre-interferometry, the lunar-occultation diffraction-fringe pattern was the highest angular-resolution measurement available. Diameters and binary status were determined for hundreds of giants and supergiants this way; some measurements (e.g., Antares's diameter, ~700 R_☉) remain among the most precise we have.
• Spacecraft target selection. Before New Horizons' 2019 flyby of Arrokoth (formerly 2014 MU69), occultation observations confirmed the object's elongated bilobate shape, allowing the spacecraft team to plan the encounter geometry months in advance. Without occultations, Arrokoth would have been a featureless point until the spacecraft arrived.

Diffraction-limited timing and the Fresnel scale

The "sharpness" of a stellar occultation ingress is set by physical optics, not the underlying body. Light from the background star, treated as a point source at infinity, encounters the foreground body's edge and diffracts. The transition from full brightness to zero occurs over a length scale called the Fresnel scale F = √(λ·d/2), where λ is the wavelength and d the distance to the occulter. For optical wavelengths (550 nm) at 1 AU, F ≈ 200 m; at 30 AU (Pluto distance), F ≈ 1.1 km. The transition has a characteristic oscillating pattern (the Fresnel fringe pattern) just before complete shadow.

For sufficiently fast cadence and sufficiently bright stars, the Fresnel fringes themselves can be resolved. They contain information about whether the limb is sharp or diffuse: an opaque atmosphere broadens them, a transparent atmospheric layer modulates their amplitude, surface topography on the Fresnel scale shifts their phase. High-speed lunar occultation photometry was historically used in this mode to measure stellar diameters with sub-mas precision.

Common pitfalls

• Inaccurate timing. The single largest source of error in occultation observations is the observer's clock. GPS-disciplined clocks give ~1 ms accuracy; computer clocks without GPS may drift seconds per night. Without sub-100 ms timing the chord lengths are too noisy to recover shape.
• Path prediction misses. Even with Gaia, ephemeris errors can shift the shadow path by tens of km. Always observe near the centre of the predicted path, allow margin, and ideally use multiple stations to maximise coverage.
• Confusing diffraction with atmosphere. A gradual ingress always indicates the line of sight is sampling something — but distinguishing a thin atmosphere from limb topography requires careful modelling. Pluto's atmosphere was confirmed by the exponential pressure profile, not just the gradual drop.
• Assuming the chord is a diameter. Each observer's chord is one chord across the silhouette, not necessarily a diameter. Inferring a diameter from a single chord overestimates the body's size if the chord misses the centre. Always use multi-station data when possible.
• Forgetting the binary star possibility. A star that is itself an unresolved binary (separation < mas) produces a stepped occultation profile rather than a sharp drop. The lunar-occultation literature has many cases where an apparent atmospheric detection was actually a binary background star.

Variants and extensions

• Lunar occultation. The Moon as occulter. Most common type; thousands per year visible from any observatory. Used historically for stellar diameters.
• Asteroid occultation. A solar-system asteroid blocking a background star. Hundreds per year. Returns asteroid sizes and shapes.
• TNO occultation. Trans-Neptunian object as occulter. Rare (~10/year for the largest TNOs) and difficult, requiring continental-scale observer networks. Returns TNO sizes, atmospheres (Pluto), rings (Chariklo, Haumea, Quaoar?), and satellites.
• Planetary occultation. A solar-system planet blocking a background star. Rare for the inner planets, more common for the giants. Sensitive to atmospheric profiles and ring systems.
• Mutual events / interplanetary occultations. One planet (or moon) occults another body. Saturn occulted Mars in 1591 (recorded by Tycho Brahe); Jupiter's moons occult each other during Jovian equinoxes. These are not strictly stellar occultations but use the same geometric framework.